Although carbohydrate antigens are important in pathogenesis of both bacterial disease and tumors, they don't provoke the same type of immune response as protein antigens. The response doesn't usually involve antigen presentation or interaction of T helper cells. Our laboratory and others have data indicating that the immune response to saccharides can be mimicked using a peptide as an antigen. This is important because the immune response to medically relevant saccharides is weak in children under two years of age, in geriatric patients, and in cancer patients. The hypothesis is that peptides could be created that would mimic a carbohydrate, with sufficient specific and with enhanced immunogencity, which could prove clinically useful in cancer immunotherapy and in the prevention of infectious disease. One carbohydrate antigen, galactose Beta 1-3 N-acetyl galactosamine, (T Ag), a tumor associated antigen was chosen to study as a model. Polyclonal and a monoclonal antibody to T ag were used to capture phage clones which bear a peptide mimic of this disaccharide. Immunization of mice with these phage clones resulted in production of antibody to T ag. The objectives of this research is to compare antibody produced to the saccharide (Abc) and antibody produced to the phage mimic (Abm) in order to determine the clinical utility of the phage mimic. Specific aims are as follows: 1) Determine the relative specificities of the two antibodies to see if Abm is of sufficient fidelity to be clinically useful. 2) Compare the characteristics of the immune response in terms of isotype, affinity constant and VH gene usage to determine if the Abm response would result in increased production of protective antibodies. 3) Analyze the biologic activity of the Abm antibodies in an assay that shows if the antibody to the mimic (Abm), like the antibody to the saccharide (Abc), can block tumor cell adhesion to endothelial cells. 4) Determine the consensus sequence of mimics of a disaccharide antigen to ascertain if such sequences can be predicted based on the sugars and linkages one wishes to mimic, to simplify future efforts to increase responsiveness to biomedically relevant carbohydrate antigens.